Remote diagnosis and maintenance guiding system of vehicle battery pack box

Abstract

The invention belongs to the technical field of battery health detection, and particularly relates to a remote diagnosis and maintenance guiding system for a vehicle battery pack box. The remote diagnosis and maintenance guidance system comprises a data acquisition edge processing module, a multi-mode communication and cache transmission module, a composite fault diagnosis module, a sensor self-calibration and environment compensation module, a cloud big data analysis and model iteration module, a remote maintenance guidance and AR visualization module and a system monitoring module. According to the invention, through multi-mode communication and edge caching, data is ensured not to be lost in an extreme environment; a composite fault is identified through a fusion diagnosis algorithm, and maintenance direction deviation is avoided; according to the invention, a sensor self-calibration and compensation mechanism is provided, false alarms caused by low temperature / high humidity are inhibited, and the accuracy of battery data monitoring is improved.

Description

Technical Field

[0001] This invention relates to the field of battery health testing technology, specifically a remote diagnostic and maintenance guidance system for a vehicle battery pack. Background Technology

[0002] The remote diagnostic and maintenance guidance system for vehicle battery packs is a core supporting technology for new energy vehicles. Its core function is to achieve precise management of battery health status through real-time data monitoring, intelligent analysis, and remote interaction.

[0003] However, the existing system still has the following technical problems:

[0004] Data transmission reliability needs to be improved: 4G / 5G communication may be interrupted in signal blind spots such as tunnels and remote mountainous areas, resulting in the loss of real-time data;

[0005] For complex faults such as voltage imbalance and abnormal internal resistance that occur simultaneously in multi-cell series systems, existing algorithms may misjudge them as a single problem, leading to deviations in the repair direction. Extreme temperatures (such as below -30℃) or high humidity environments may cause inaccurate sensor readings and trigger false alarms.

[0006] To address these issues, a remote diagnostic and maintenance guidance system for vehicle battery packs is proposed. Summary of the Invention

[0007] The purpose of this invention is to provide a remote diagnostic and maintenance guidance system for a vehicle battery pack, in order to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a remote diagnostic and maintenance guidance system for a vehicle battery pack, the remote diagnostic and maintenance guidance system comprising a data acquisition edge processing module, a multimodal communication and cache transmission module, a composite fault diagnosis module, a sensor self-calibration and environmental compensation module, a cloud big data analysis and model iteration module, a remote maintenance guidance and AR visualization module, and a system monitoring module;

[0009] The data acquisition edge processing module acquires the voltage, current, temperature, internal resistance, and SOC / SOH parameters of the battery pack, and performs preliminary data filtering, compression, and outlier marking at the battery management system end.

[0010] The multimodal communication and cache transmission module supports multiple communication methods, such as 4G / 5G, NB-IoT, LoRa, satellite communication and V2X vehicle networking. When the signal is interrupted, it automatically switches to a low bandwidth / latency tolerance mode and enables local data caching.

[0011] The composite fault diagnosis module is based on multi-parameter coupling analysis to identify composite faults of voltage imbalance and internal resistance abnormality. It uses a fusion diagnosis algorithm to avoid misjudging composite problems as single faults.

[0012] The sensor self-calibration and environmental compensation module monitors the ambient temperature, humidity, and vibration conditions in real time, and dynamically compensates the sensor readings to suppress drift.

[0013] The cloud-based big data analysis and model iteration module aggregates data from multiple vehicles, performs global fault mode mining, and regularly updates the edge diagnostic model.

[0014] The remote maintenance guidance and AR visualization module generates a visual maintenance guidance plan based on the diagnostic results, and supports AR glasses or mobile devices to display disassembly steps and fault location.

[0015] The system monitoring module encrypts data transmission and monitors and alerts on the system's operating status.

[0016] Preferably, the data acquisition edge processing module includes a multi-parameter sensor acquisition unit, a signal conditioning and filtering unit, a data preprocessing and compression unit, a local fault initial judgment and alarm unit, an edge cache breakpoint resume unit, a time synchronization data marking unit, and an environmental monitoring status perception unit.

[0017] The multi-parameter sensor acquisition unit is used to acquire the individual cell voltage, total voltage, total current, cell temperature, internal resistance, SOC, and SOH of each cell or module in the battery pack;

[0018] The signal conditioning and filtering unit is used to amplify, denoise, and perform anti-interference processing on the original analog signal, and uses digital filtering algorithms to remove high-frequency noise.

[0019] The data preprocessing and compression unit is used to remove obvious outliers, unify data of different dimensions to the standard range, and reduce data volume by using Delta encoding and LZ77 algorithm.

[0020] The local fault preliminary judgment and alarm unit performs preliminary fault identification based on preset rules or a lightweight model;

[0021] The edge cache breakpoint resume transmission unit temporarily stores data in local storage when 4G / 5G signal is interrupted. After communication is restored, the cached data is retransmitted in chronological order, and priority transmission is supported.

[0022] The time synchronization data marking unit uses GPS or Beidou time synchronization to ensure that all collected data has a high-precision timestamp and marks the data source, vehicle VIN code, and geographical location;

[0023] The environmental monitoring status sensing unit is used to collect external environmental parameters of the Pack box.

[0024] Preferably, the multimodal communication and cache transmission module includes a multi-channel communication access unit, a communication switching routing unit, an edge data cache queue management unit, a breakpoint resume data transmission unit, a data compression encoding optimization unit, a communication status monitoring and alarm unit, and a security encryption and identity authentication unit.

[0025] The multi-channel communication access unit integrates multiple communication interfaces; the communication switching routing unit evaluates the signal strength, latency, bandwidth, power consumption, and operator tariff policies of each communication channel in real time; the edge data cache queue management unit establishes a multi-level cache queue in local memory; the breakpoint resume data transmission unit records the offset or timestamp of successfully uploaded data, and resumes uploading untransmitted data from the breakpoint after communication is restored, supporting data deduplication and reordering to ensure complete and orderly cloud reception; the data compression encoding optimization unit is used to compress the data to be transmitted; the communication status monitoring and alarm unit monitors the connection status, signal quality, data throughput, and number of failures of each communication channel in real time; and the security encryption and authentication unit encrypts data transmission.

[0026] Preferably, the composite fault diagnosis module includes a multi-parameter coupling analysis unit, a fault feature extraction and representation unit, a fusion diagnosis algorithm engine unit, a fault association map construction unit, and a fault confidence assessment and decision-making unit;

[0027] The multi-parameter coupling analysis unit analyzes the voltage consistency, internal resistance variation trend, temperature distribution uniformity, and SOC / SOH differences of multiple cells in real time; the fault feature extraction and representation unit extracts key features from time-series data; the fusion diagnostic algorithm engine unit integrates multiple diagnostic algorithms to achieve complementary advantages; the fault association graph construction unit constructs a battery fault knowledge graph; the fault confidence assessment and decision-making unit outputs a confidence score for each diagnostic result. When the confidence score is lower than the threshold, a manual review state is triggered, generating auxiliary judgment suggestions and supporting multi-level alarms.

[0028] Preferably, the sensor self-calibration and environmental compensation module includes an environmental perception and monitoring unit, a sensor health status detection unit, a dynamic zero-point calibration offset compensation unit, a multi-source data fusion redundancy verification unit, and an environmental compensation algorithm engine unit. The environmental perception and monitoring unit deploys dedicated environmental sensors to collect environmental data inside and outside the battery pack in real time. The sensor health status detection unit monitors the output stability of each sensor, the degree of deviation from historical data, and communication integrity in real time, and uses statistical methods or lightweight models to identify abnormal trends. The dynamic zero-point calibration offset compensation unit triggers automatic calibration under specific conditions (such as static, fully charged, and unloaded) and supports periodic self-calibration. The multi-source data fusion redundancy verification unit deploys multiple sensors for the same parameter, uses a fusion algorithm to process multiple signals, and automatically downweights or removes a sensor when its data deviates significantly. The environmental compensation algorithm engine unit establishes an environment-reading mapping model to dynamically correct the original data.

[0029] Preferably, the cloud-based big data analysis and model iteration module includes a multi-source data aggregation and storage unit, a global health status assessment unit, a fault mode mining and clustering analysis unit, a diagnostic model training and optimization unit, and a model version management and OTA distribution unit. The multi-source data aggregation and storage unit accesses real-time status data, fault alarm logs, maintenance records, and environmental information from various vehicle edge devices, and uses a distributed storage system for massive data storage, supporting mixed storage of structured, semi-structured, and time-series data. The global health status assessment unit evaluates the battery SOH degradation trend of different vehicle models, batches, and regions based on big data statistical methods. The battery performance degradation rate distribution is used to generate a battery health map and identify high-risk vehicle groups. The fault mode mining and clustering analysis unit uses unsupervised learning algorithms to cluster historical fault data and identify new composite fault modes, regional faults, and batch defects. The diagnostic model training and optimization unit trains based on massive labeled data, uses cross-validation and A / B testing to evaluate model performance, supports federated learning, and jointly trains models without obtaining the original data. The model version management and OTA distribution unit establishes a model repository, records model version numbers, training datasets, performance indicators, applicable scenarios, supports OTA remote push of new models to vehicle edge devices, and supports canary release and rollback mechanisms.

[0030] Preferably, the remote maintenance guidance and AR visualization module includes a maintenance task generation and scheduling unit, a 3D battery pack digital twin model unit, an AR augmented reality rendering engine unit, a multi-terminal adaptation and interaction unit, and a remote expert collaboration and annotation unit. The maintenance task generation and scheduling unit receives diagnostic results from the composite fault diagnosis module and automatically generates structured maintenance tasks. The 3D battery pack digital twin model unit constructs a high-precision 3D model consistent with the physical battery pack, and the model is bound to the actual vehicle VIN, supporting vehicle customization. The AR augmented reality rendering engine unit overlays maintenance information onto the real battery pack in an AR manner. The multi-terminal adaptation and interaction unit supports multiple terminal devices and provides multiple interaction methods. The remote expert collaboration and annotation unit supports remote experts accessing the maintenance site via video call, supporting multi-person collaborative consultation.

[0031] Preferably, the system monitoring module includes a device operation status monitoring unit, a communication link monitoring unit, a data flow monitoring and integrity verification unit, and a functional module health detection unit. The device operation status monitoring unit collects data on the CPU usage, memory usage, storage space, device temperature, and power supply voltage of edge devices in real time, and sets thresholds to trigger early warnings. The communication link monitoring unit monitors each communication channel in real time and records the communication interruption time, frequency, and duration. The data flow monitoring and integrity verification unit tracks the entire process of data collection, edge processing, transmission, and cloud reception, and detects data loss, duplication, and out-of-order issues. The functional module health detection unit performs heartbeat detection and service availability verification on each core module of the system, supporting health checks under a microservice architecture.

[0032] Compared with the prior art, the beneficial effects of the present invention are:

[0033] 1) Improved communication robustness: This application ensures that data is not lost in extreme environments through multimodal communication and edge caching;

[0034] 2) Enhanced diagnostic accuracy: This application identifies complex faults through a fusion diagnostic algorithm, avoiding deviations in maintenance direction;

[0035] 3) Environmental adaptability optimization: This application has a sensor self-calibration and compensation mechanism to suppress false alarms caused by low temperature / high humidity and increase the accuracy of battery data monitoring. Attached Figure Description

[0036] Figure 1 This is a diagram showing the module composition of this system;

[0037] Figure 2 This is a diagram showing the unit composition of the data acquisition and edge processing module.

[0038] Figure 3This is a diagram showing the unit composition of the multimodal communication and buffered transmission module;

[0039] Figure 4 This is a diagram showing the unit composition of the composite fault diagnosis module;

[0040] Figure 5 This is a diagram showing the unit composition of the sensor self-calibration and environmental compensation module.

[0041] Figure 6 This is a diagram showing the unit composition of the cloud-based big data analysis and model iteration module.

[0042] Figure 7 A diagram showing the unit composition of the remote maintenance guidance and AR visualization module;

[0043] Figure 8 This is a diagram showing the unit composition of the system monitoring module. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Example:

[0046] Please see Figure 1-8 The present invention provides a technical solution:

[0047] A remote diagnostic and maintenance guidance system for a vehicle battery pack includes a data acquisition and edge processing module, a multimodal communication and cache transmission module, a composite fault diagnosis module, a sensor self-calibration and environmental compensation module, a cloud-based big data analysis and model iteration module, a remote maintenance guidance and AR visualization module, and a system monitoring module.

[0048] The data acquisition edge processing module acquires the voltage, current, temperature, internal resistance, and SOC / SOH parameters of the battery pack, and performs preliminary data filtering, compression, and outlier marking at the battery management system end.

[0049] The multimodal communication and cache transmission module supports multiple communication methods, such as 4G / 5G, NB-IoT, LoRa, satellite communication and V2X vehicle networking. When the signal is interrupted, it automatically switches to a low bandwidth / latency tolerance mode and enables local data caching.

[0050] The composite fault diagnosis module is based on multi-parameter coupling analysis to identify composite faults of voltage imbalance and internal resistance abnormality. It uses a fusion diagnosis algorithm to avoid misjudging composite problems as single faults.

[0051] The sensor self-calibration and environmental compensation module monitors the ambient temperature, humidity, and vibration conditions in real time, and dynamically compensates the sensor readings to suppress drift.

[0052] The cloud-based big data analysis and model iteration module aggregates data from multiple vehicles, performs global fault mode mining, and regularly updates the edge diagnostic model.

[0053] The remote maintenance guidance and AR visualization module generates a visual maintenance guidance plan based on the diagnostic results, and supports AR glasses or mobile devices to display disassembly steps and fault location.

[0054] The system monitoring module encrypts data transmission and monitors and alerts on the system's operating status.

[0055] The data acquisition edge processing module includes a multi-parameter sensor acquisition unit, a signal conditioning and filtering unit, a data preprocessing and compression unit, a local fault initial judgment and alarm unit, an edge cache breakpoint resume unit, a time synchronization data marking unit, and an environmental monitoring status perception unit.

[0056] The multi-parameter sensor acquisition unit is used to acquire the individual cell voltage (mV level accuracy), total voltage, total current (ampere level), cell temperature (multi-point NTC or infrared temperature measurement), internal resistance (estimated by pulse discharge method or AC impedance method), SOC (state of charge), and SOH (state of health) of each cell or module in the battery pack, providing raw data input, which is the basis for subsequent diagnosis. High-frequency sampling (such as 10Hz-100Hz) ensures the capture of transient anomalies.

[0057] The signal conditioning and filtering unit is used to amplify, denoise, and perform anti-interference processing on the original analog signal. It uses digital filtering algorithms to remove high-frequency noise, improve the signal-to-noise ratio, prevent data distortion caused by electromagnetic interference, and avoid false alarms or diagnostic deviations caused by noise.

[0058] The data preprocessing and compression unit is used to remove obvious outliers, unify data of different dimensions to the standard range, reduce data volume by using Delta encoding and LZ77 algorithm, reduce communication bandwidth requirements, improve transmission efficiency, reduce cloud processing pressure, and improve the overall system response speed.

[0059] The local fault preliminary judgment and alarm unit performs preliminary fault identification based on preset rules or lightweight models, realizes edge intelligence, can still provide early warnings when there is no network, quickly respond to emergency faults, and improve security.

[0060] The edge cache breakpoint resume unit temporarily stores data in local storage when 4G / 5G signals are interrupted. After communication is restored, the cached data is retransmitted in chronological order. It supports priority transmission to ensure that critical data is not permanently lost due to communication interruption and supports historical data backtracking analysis.

[0061] The time synchronization data marking unit uses GPS or Beidou time synchronization to ensure that all collected data has a high-precision timestamp, marking the data source, vehicle VIN code, and geographical location, supporting data alignment and correlation analysis of multiple vehicles and multiple time points, which facilitates fault tracing and responsibility determination.

[0062] The environmental monitoring and state sensing unit is used to collect external environmental parameters of the Pack box, including ambient temperature (-40℃-85℃), relative humidity (0-100%RH), and vibration acceleration; to provide a basis for subsequent sensor drift compensation and to determine whether extreme environments affect battery performance.

[0063] The multimodal communication and cache transmission module includes a multi-channel communication access unit, a communication switching routing unit, an edge data cache queue management unit, a breakpoint resume data transmission unit, a data compression encoding optimization unit, a communication status monitoring and alarm unit, and a security encryption and identity authentication unit.

[0064] The multi-channel communication access unit integrates various communication interfaces, including cellular networks, narrowband IoT, long-range low-power communication, satellite communication, and vehicle-to-everything (V2X) communication. In areas with insufficient 4G / 5G coverage, such as tunnels, mountainous areas, and Gobi deserts, it automatically switches to LoRa or satellite communication, providing communication redundancy and improving system robustness. The communication switching routing unit evaluates the signal strength, latency, bandwidth, power consumption, and operator pricing policies of each communication channel in real time to achieve seamless switching, avoid data loss due to communication interruptions, optimize resource utilization, and reduce operating costs. The edge data cache queue management unit establishes a multi-level cache queue in local storage, including high-priority, medium-priority, and low-priority queues. The breakpoint resume data transmission unit records the offset or timestamp of successfully uploaded data. After communication is restored, it continues to upload untransmitted data from the breakpoint, supporting data deduplication and reordering to ensure complete and orderly cloud reception. The data compression encoding optimization unit compresses the data to be transmitted. The communication status monitoring and alarm unit monitors the connection status, signal quality, data throughput, and number of failures of each communication channel in real time. The security encryption and authentication unit encrypts data transmission.

[0065] The composite fault diagnosis module includes a multi-parameter coupling analysis unit, a fault feature extraction and representation unit, a fusion diagnosis algorithm engine unit, a fault association graph construction unit, and a fault confidence assessment and decision unit. The multi-parameter coupling analysis unit analyzes the voltage consistency, internal resistance variation trend, temperature distribution uniformity, and SOC / SOH differences of multiple cells in real time, distinguishing between simple voltage imbalance and composite anomalies caused by aging or poor contact, avoiding misjudging multi-factor problems as single faults. The fault feature extraction and representation unit extracts key features from time-series data. The fusion diagnosis algorithm engine unit integrates multiple diagnosis algorithms to achieve complementary advantages. The fault association graph construction unit constructs a battery fault knowledge graph. The fault confidence assessment and decision unit outputs a confidence score for each diagnostic result; when the confidence score is below a threshold, a manual review state is triggered, generating auxiliary judgment suggestions and supporting multi-level alarms.

[0066] The sensor self-calibration and environmental compensation module includes an environmental perception and monitoring unit, a sensor health status detection unit, a dynamic zero-point calibration offset compensation unit, a multi-source data fusion redundancy verification unit, and an environmental compensation algorithm engine unit. The environmental perception and monitoring unit deploys dedicated environmental sensors, including a temperature sensor (supporting -40℃ to +85℃), a humidity sensor (0-100% RH), a barometer, and a triaxial accelerometer, to collect real-time environmental data inside and outside the battery pack, providing contextual input for subsequent compensation algorithms and determining whether extreme operating conditions are present. The sensor health status detection unit monitors the output of each sensor in real-time. The system assesses stability, deviation from historical data, and communication integrity, providing early warnings of sensor aging, poor contact, or damage to prevent data distortion caused by malfunctions. It uses statistical methods or lightweight models to identify abnormal trends. The dynamic zero-point calibration offset compensation unit triggers automatic calibration under specific conditions (such as static, fully charged, or unloaded) and supports periodic self-calibration. The multi-source data fusion redundancy verification unit deploys multiple sensors for the same parameter, using a fusion algorithm to process multiple signals. When a sensor's data deviates significantly, it automatically downweights or removes it. The environmental compensation algorithm engine unit establishes an environment-reading mapping model to dynamically correct the original data.

[0067] The cloud-based big data analysis and model iteration module includes a multi-source data aggregation and storage unit, a global health status assessment unit, a fault mode mining and clustering analysis unit, a diagnostic model training and optimization unit, and a model version management and OTA distribution unit. The multi-source data aggregation and storage unit accesses real-time status data, fault alarm logs, maintenance records, and environmental information from various vehicle edge devices, and uses a distributed storage system for massive data storage, supporting mixed storage of structured, semi-structured, and time-series data. The global health status assessment unit evaluates the battery SOH degradation trend of different models, batches, and regions based on big data statistical methods. The battery performance degradation rate distribution is used to generate a battery health map and identify high-risk vehicle groups. The fault mode mining and clustering analysis unit uses unsupervised learning algorithms to cluster historical fault data and identify new composite fault modes, regional faults, and batch defects. The diagnostic model training and optimization unit trains based on massive labeled data, uses cross-validation and A / B testing to evaluate model performance, supports federated learning, and jointly trains models without obtaining the original data. The model version management and OTA distribution unit establishes a model repository, records model version numbers, training datasets, performance indicators, applicable scenarios, supports OTA remote push of new models to vehicle edge devices, and supports canary release and rollback mechanisms.

[0068] The remote maintenance guidance and AR visualization module includes a maintenance task generation and scheduling unit, a 3D battery pack digital twin model unit, an AR augmented reality rendering engine unit, a multi-terminal adaptation and interaction unit, and a remote expert collaboration and annotation unit. The maintenance task generation and scheduling unit receives diagnostic results from the composite fault diagnosis module and automatically generates structured maintenance tasks. The 3D battery pack digital twin model unit constructs a high-precision 3D model consistent with the physical battery pack, and the model is bound to the actual vehicle's VIN, supporting vehicle model customization. The AR augmented reality rendering engine unit overlays maintenance information onto the real battery pack using AR technology. The multi-terminal adaptation and interaction unit supports various terminal devices and provides multiple interaction methods. The remote expert collaboration and annotation unit allows remote experts to access the maintenance site via video call, supporting multi-person collaborative consultations.

[0069] The system monitoring module includes a device operation status monitoring unit, a communication link monitoring unit, a data flow monitoring and integrity verification unit, and a functional module health detection unit. The device operation status monitoring unit collects real-time data on edge device CPU usage, memory consumption, storage space, device temperature, and power supply voltage, sets thresholds to trigger early warnings, and promptly identifies edge computing resource bottlenecks to prevent data interruptions due to device crashes. The communication link monitoring unit monitors each communication channel in real-time, records communication interruption time, frequency, and duration, identifies high-incidence areas of signal blind spots, and provides decision-making basis for multimodal communication switching. The data flow monitoring and integrity verification unit tracks the entire data process from acquisition, edge processing, transmission, to cloud reception, detects data loss, duplication, and out-of-order issues, identifies data gaps caused by buffer overflows or transmission failures, and supports automatic retransmission mechanisms. The functional module health detection unit performs heartbeat detection and service availability verification on each core module of the system, supports health checks under a microservice architecture, quickly locates faulty modules, avoids black-box maintenance, and supports automatic restart or switching to backup services.

[0070] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered exemplary and non-limiting in all respects. The scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.

[0071] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A remote diagnostic and maintenance guidance system for a vehicle battery pack, characterized in that, The remote diagnostic and maintenance guidance system includes a data acquisition and edge processing module, a multimodal communication and buffer transmission module, a composite fault diagnosis module, a sensor self-calibration and environmental compensation module, a cloud-based big data analysis and model iteration module, a remote maintenance guidance and AR visualization module, and a system monitoring module. The data acquisition edge processing module acquires the voltage, current, temperature, internal resistance, and SOC / SOH parameters of the battery pack, and performs preliminary data filtering, compression, and outlier marking at the battery management system end. The multimodal communication and cache transmission module supports multiple communication methods, such as 4G / 5G, NB-IoT, LoRa, satellite communication and V2X vehicle networking. When the signal is interrupted, it automatically switches to a low bandwidth / latency tolerance mode and enables local data caching. The composite fault diagnosis module is based on multi-parameter coupling analysis to identify composite faults of voltage imbalance and internal resistance abnormality. It uses a fusion diagnosis algorithm to avoid misjudging composite problems as single faults. The sensor self-calibration and environmental compensation module monitors the ambient temperature, humidity, and vibration conditions in real time, and dynamically compensates the sensor readings to suppress drift. The cloud-based big data analysis and model iteration module aggregates data from multiple vehicles, performs global fault mode mining, and regularly updates the edge diagnostic model. The remote maintenance guidance and AR visualization module generates a visual maintenance guidance plan based on the diagnostic results, and supports AR glasses or mobile devices to display disassembly steps and fault location. The system monitoring module encrypts data transmission and monitors and alerts on the system's operating status.

2. The remote diagnostic and maintenance guidance system for a vehicle battery pack according to claim 1, wherein the data acquisition edge processing module includes a multi-parameter sensor acquisition unit, a signal conditioning and filtering unit, a data preprocessing and compression unit, a local fault preliminary judgment and alarm unit, an edge cache breakpoint resume transmission unit, a time synchronization data marking unit, and an environmental monitoring status perception unit. The multi-parameter sensor acquisition unit is used to acquire the individual cell voltage, total voltage, total current, cell temperature, internal resistance, SOC, and SOH of each cell or module in the battery pack; The signal conditioning and filtering unit is used to amplify, denoise, and perform anti-interference processing on the original analog signal, and uses digital filtering algorithms to remove high-frequency noise. The data preprocessing and compression unit is used to remove obvious outliers, unify data of different dimensions to the standard range, and reduce data volume by using Delta encoding and LZ77 algorithm. The local fault preliminary judgment and alarm unit performs preliminary fault identification based on preset rules or a lightweight model; The edge cache breakpoint resume transmission unit temporarily stores data in local storage when 4G / 5G signal is interrupted. After communication is restored, the cached data is retransmitted in chronological order, and priority transmission is supported. The time synchronization data marking unit uses GPS or Beidou time synchronization to ensure that all collected data has a high-precision timestamp and marks the data source, vehicle VIN code, and geographical location; The environmental monitoring status sensing unit is used to collect external environmental parameters of the Pack box.

3. The remote diagnostic and maintenance guidance system for a vehicle battery pack according to claim 1, characterized in that: The multimodal communication and cache transmission module includes a multi-channel communication access unit, a communication switching routing unit, an edge data cache queue management unit, a breakpoint resume data transmission unit, a data compression encoding optimization unit, a communication status monitoring and alarm unit, and a security encryption and identity authentication unit. The multi-channel communication access unit integrates multiple communication interfaces; the communication switching routing unit evaluates the signal strength, latency, bandwidth, power consumption, and operator tariff policies of each communication channel in real time; the edge data cache queue management unit establishes a multi-level cache queue in local memory; the breakpoint resume data transmission unit records the offset or timestamp of successfully uploaded data, and resumes uploading untransmitted data from the breakpoint after communication is restored, supporting data deduplication and reordering to ensure complete and orderly cloud reception; the data compression encoding optimization unit is used to compress the data to be transmitted; the communication status monitoring and alarm unit monitors the connection status, signal quality, data throughput, and number of failures of each communication channel in real time; and the security encryption and authentication unit encrypts data transmission.

4. The remote diagnostic and maintenance guidance system for a vehicle battery pack according to claim 1, characterized in that: The composite fault diagnosis module includes a multi-parameter coupling analysis unit, a fault feature extraction and representation unit, a fusion diagnosis algorithm engine unit, a fault association map construction unit, and a fault confidence assessment and decision-making unit. The multi-parameter coupling analysis unit analyzes the voltage consistency, internal resistance variation trend, temperature distribution uniformity, and SOC / SOH differences of multiple cells in real time; the fault feature extraction and representation unit extracts key features from time-series data; the fusion diagnostic algorithm engine unit integrates multiple diagnostic algorithms to achieve complementary advantages; the fault association graph construction unit constructs a battery fault knowledge graph; the fault confidence assessment and decision-making unit outputs a confidence score for each diagnostic result. When the confidence score is lower than the threshold, a manual review state is triggered, generating auxiliary judgment suggestions and supporting multi-level alarms.

5. The remote diagnostic and maintenance guidance system for a vehicle battery pack according to claim 1, characterized in that: The sensor self-calibration and environmental compensation module includes an environmental sensing and monitoring unit, a sensor health status detection unit, a dynamic zero-point calibration offset compensation unit, a multi-source data fusion redundancy verification unit, and an environmental compensation algorithm engine unit. The environmental sensing and monitoring unit deploys dedicated environmental sensors to collect real-time environmental data inside and outside the battery pack. The sensor health status detection unit monitors the output stability of each sensor, the degree of deviation from historical data, and communication integrity in real time, using statistical methods or lightweight models to identify abnormal trends. The dynamic zero-point calibration offset compensation unit triggers automatic calibration under specific conditions (such as static, fully charged, and unloaded), supporting periodic self-calibration. The multi-source data fusion redundancy verification unit deploys multiple sensors for the same parameter, uses a fusion algorithm to process multiple signals, and automatically downweights or removes sensors when their data deviates significantly. The environmental compensation algorithm engine unit establishes an environment-reading mapping model to dynamically correct the original data.

6. The remote diagnostic and maintenance guidance system for a vehicle battery pack according to claim 1, characterized in that: The cloud-based big data analysis and model iteration module includes a multi-source data aggregation and storage unit, a global health status assessment unit, a fault mode mining and clustering analysis unit, a diagnostic model training and optimization unit, and a model version management and OTA distribution unit. The multi-source data aggregation and storage unit accesses real-time status data, fault alarm logs, maintenance records, and environmental information from various vehicle edge devices, and uses a distributed storage system for massive data storage, supporting mixed storage of structured, semi-structured, and time-series data. The global health status assessment unit evaluates the battery SOH degradation trend and battery performance degradation rate distribution for different vehicle models, batches, and regions based on big data statistical methods, generates a battery health map, and identifies high-risk vehicle groups. The fault mode mining and clustering analysis unit uses unsupervised learning algorithms to cluster historical fault data, identifying novel composite fault modes, regional faults, and batch defects. The diagnostic model training and optimization unit trains based on massive labeled data, uses cross-validation and A / B testing to evaluate model performance, supports federated learning, and jointly trains models without obtaining the original data. The model version management and OTA distribution unit establishes a model repository, records model version numbers, training datasets, performance metrics, applicable scenarios, supports OTA remote push of new models to vehicle edge devices, and supports canary release and rollback mechanisms.

7. The remote diagnostic and maintenance guidance system for a vehicle battery pack according to claim 1, characterized in that: The remote maintenance guidance and AR visualization module includes a maintenance task generation and scheduling unit, a 3D battery pack digital twin model unit, an AR augmented reality rendering engine unit, a multi-terminal adaptation and interaction unit, and a remote expert collaboration and annotation unit. The maintenance task generation and scheduling unit receives diagnostic results from the composite fault diagnosis module and automatically generates structured maintenance tasks. The 3D battery pack digital twin model unit constructs a high-precision 3D model consistent with the physical battery pack, and the model is bound to the actual vehicle's VIN, supporting vehicle model customization. The AR augmented reality rendering engine unit overlays maintenance information onto the real battery pack using AR technology. The multi-terminal adaptation and interaction unit supports various terminal devices and provides multiple interaction methods. The remote expert collaboration and annotation unit allows remote experts to access the maintenance site via video call, supporting multi-person collaborative consultations.

8. The remote diagnostic and maintenance guidance system for a vehicle battery pack according to claim 1, characterized in that: The system monitoring module includes a device operation status monitoring unit, a communication link monitoring unit, a data flow monitoring and integrity verification unit, and a functional module health detection unit; the device operation status monitoring unit collects the CPU usage, memory usage, storage space, device temperature, and power supply voltage of the edge device in real time, and sets thresholds to trigger early warnings; The communication link monitoring unit monitors each communication channel in real time and records the communication interruption time, frequency, and duration. The data flow monitoring and integrity verification unit tracks the entire process of data acquisition, edge processing, transmission, and cloud reception, and detects data loss, duplication, and out-of-order issues. The functional module health detection unit performs heartbeat detection and service availability verification on each core module of the system, supporting health checks under a microservice architecture.

Images